Life cycle inventory

The life cycle inventory (LCI) is the list of every elementary flow crossing the system boundary (resources taken from nature, emissions released to air, water and soil), per functional unit.

Unit processes

The building block of an inventory is the unit process: the smallest part of the system for which you have input and output data. For each unit process you record:

• Product flows (technosphere): the goods and services it produces and consumes, such as 1 kg of steel sheet produced from 0.6 kWh of electricity and 1.1 kg of iron ore.
• Elementary flows (biosphere): exchanges with nature, such as 1.7 kg of fossil CO₂ to air.

Link unit processes through their product flows and you have a product system. Scale each one to the amount the functional unit needs and add up the elementary flows: that is the inventory. Lesson 8 does exactly this by hand, and LCA software does it with matrices.

Foreground and background

• Foreground processes are the ones you (or your client) control: your factory, your bill of materials, your transport. Collect primary data here: meter readings, purchase records, production logs.
• Background processes are the supply chain you don’t control: grid electricity, commodity steel, diesel. Use secondary data from LCI databases here.

A good study spends its data-collection effort where it changes the answer. A screening run with secondary data first, then primary data for the hotspots, is standard practice.

Collecting foreground data

For each foreground process, gather, over a representative period (usually one year):

• Production output (in the reference unit)
• Energy use by carrier: electricity, gas, diesel, steam
• Material inputs from the bill of materials, including packaging and scrap rates
• Direct emissions (from permits, measurements or calculation)
• Waste outputs and how each is treated
• Transport distances and modes for inputs

Tip

Always ask for the scrap rate. If a factory buys 1.15 kg of steel for every 1 kg that ends up in the product, the missing 0.15 kg is a real input, and forgetting it is one of the most common inventory errors.

Data quality

ISO 14044 asks you to describe data quality against requirements set in the scope:

Aspect	Question
Time-related coverage	How old are the data?
Geographical coverage	Do they represent the right country or region?
Technology coverage	Do they represent the actual technology, or an average?
Precision	How variable are the values?
Completeness	What share of flows is measured, not estimated?
Representativeness	Overall, do the data reflect the system you’re studying?
Consistency and reproducibility	Were methods applied uniformly, and could someone repeat it?
Sources and uncertainty	Where did each value come from, and how uncertain is it?

Record the source of every number as you go. A spreadsheet column for “source” costs nothing now and saves days during review.

Background databases

Background data come from LCI databases. Some are free (e.g. USLCI, Agribalyse), some are licensed (e.g. ecoinvent, Sphera’s managed LCA content). They differ in coverage, in system model (how they handle allocation and recycling), and in how often they’re updated. Mixing databases is possible but needs care: see the data and tools page.

What you should now be able to do

• Describe a unit process as product flows plus elementary flows.
• Decide where you need primary data and where secondary data will do.
• Write a data-collection request for a supplier’s factory.